Machine Learning Engineer

Chagible; Singapore; Remoto; SGD 90.000 - 170.000; Tempo parcial

Resumo da vaga

Chagible AI Lab seeks a Machine Learning Engineer to design, develop, and deploy ML models, with a focus on large language models and production-ready systems. You will fine-tune LLMs, build data pipelines, and implement prompt engineering while optimizing inference performance and cost.

The role involves experimenting to improve outputs, collaborating with annotators and QA teams, and integrating models into web apps and internal tools. Remote freelance setup with flexible hours.

Qualificações

• Strong proficiency in Python and hands-on PyTorch or TensorFlow
• Solid understanding of deep learning fundamentals and optimization
• Knowledge of transformer architectures and LLM internals
• Experience fine-tuning LLMs (LoRA, PEFT, SFT, instruction tuning)
• Experience with retrieval-augmented generation and embeddings
• Experience with vector databases (Pinecone, Weaviate, FAISS)
• Experience building evaluation frameworks for LLM outputs
• Familiarity with prompt engineering and structured output control
• Experience deploying models via APIs and optimizing latency/cost
• Knowledge of distributed training and GPU utilization
• Experience handling large-scale datasets for training/evaluation
• Strong debugging and experimental design skills

Responsabilidades

• Design, develop, and deploy ML models for real-world applications
• Fine-tune and optimize LLMs for performance
• Build and maintain data pipelines for training and evaluation
• Implement prompt engineering strategies and evaluation frameworks
• Conduct experiments to improve model outputs and reliability
• Monitor model performance and iterate improvements
• Collaborate with data annotators and QA teams on training data
• Optimize inference speed, scalability, and cost efficiency
• Integrate models into web apps and internal tools
• Document model architecture and technical decisions

Conhecimentos

Python

Deep learning

Transformer models

LLMs

Retrieval augmented generation

Embeddings

Prompt engineering

Model deployment

Distributed training

GPUs

Experiment design

Model evaluation

Ferramentas

PyTorch

TensorFlow

Pinecone

Weaviate

FAISS

Descrição da vaga

We are looking for a Machine Learning Engineer who is analytical, innovative, and deeply curious about how intelligent systems learn and improve. If you enjoy building, fine-tuning, and optimizing AI models, especially large language models, and turning research into production-ready systems, you will thrive in this role.

What You Will Do

As our Machine Learning Engineer, you will:

• Design, develop, and deploy machine learning models for real-world applications
• Fine-tune and optimize large language models (LLMs) for performance and accuracy
• Build and maintain data pipelines for model training and evaluation
• Implement prompt engineering strategies and LLM evaluation frameworks
• Conduct experiments to improve model outputs, reliability, and efficiency
• Monitor model performance and implement continuous improvements
• Collaborate with data annotators and QA teams to refine training data
• Optimize inference speed, scalability, and cost efficiency
• Integrate AI models into web applications and internal tools
• Document model architecture, workflows, and technical decisions

What We Are Looking For

You might be a great fit if you have:

• Strong proficiency in Python with hands‑on experience in PyTorch or TensorFlow
• Solid understanding of deep learning fundamentals, including backpropagation, optimization algorithms, and loss functions
• Strong knowledge of transformer architectures, attention mechanisms, and large language model internals
• Experience fine‑tuning LLMs using techniques such as LoRA, PEFT, instruction tuning, or supervised fine‑tuning (SFT)
• Hands‑on experience designing and implementing retrieval‑augmented generation (RAG) systems, including embeddings, vector databases, and retrieval pipelines
• Experience working with embeddings and vector databases (e.g., Pinecone, Weaviate, FAISS) for semantic search and retrieval
• Experience building evaluation frameworks for LLM outputs, including hallucination detection and response scoring
• Familiarity with prompt engineering strategies and structured output control
• Experience deploying models via APIs and optimizing inference performance (latency, throughput, and cost efficiency)
• Knowledge of distributed training, GPU utilization, and model optimization techniques
• Experience handling large‑scale structured and unstructured datasets for training and evaluation
• Strong debugging skills for model behavior analysis and performance bottlenecks
• Ability to design experiments, interpret metrics, and iterate based on empirical results

Your Work Setup

• Remote freelance work
• 4 days per week, 12:00 AM to 9:00 AM (UTC+8)
• Stable workload with structured experimentation and development cycles

At the moment, we are not hiring applicants located outside the specified countries.